Regulation of Reactive Oxygen Species by Atm Is Essential for Proper Response to DNA Double-Strand Breaks in Lymphocytes

Mice

Atm^+/− mice (C57BL/6) were a gift from P. J. McKinnon (St. Jude Children’s Research Hospital, Memphis, TN). Offspring of intercrosses were genotyped using PCR-based assays of tail DNA. Littermates were used as controls in all experiments. Animal care in our laboratory was in accordance with the guidelines for animal and recombinant DNA experiments in Keio University and Kanazawa University. For in vivo treatment, NAC in drinking water (1 mg/ml) was administered to WT or Atm^−/− mice.

Generation and culture of primary MEFs

Primary MEFs were derived from WT or Atm^−/− embryos at E14.5. MEFs were cultured in DMEM supplemented with 10% FCS, 5 mM glutamine, 50 μM 2-ME, 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C in a humidified atmosphere containing 5% CO_2. For NAC experiments, MEFs were incubated with 100 μM NAC (Sigma-Aldrich). SA-β-gal staining was performed according to the manufacturer’s protocol (Cell Signaling Technology).

Flow cytometry

The following mAbs purchased from BD Pharmingen were used for flow cytometric analyses: Thy1 (53-2.1), CD3, CD4 (L3T4), CD8 (53-6.72), B220 (RA3-6B2), CD43, IgM, IgG1, TCRαβ, or TCRγδ.

Ab detection by ELISA

Total IgA, IgG1, IgG2a, IgG3, IgM, and IgE were measured using an Ig Detection Kit purchased from Southern Biotechnology Associates, as described previously (18).

Cell cycle analysis

G₁ arrest was examined in murine embryonic fibroblasts (MEFs) as described previously (19). MEFs were synchronized at G₀ by incubation for 4 days in DMEM containing 0.1% serum. The G₀-synchronized cells were trypsinized and resuspended in growth medium containing 65 μM BrdU. The cells were irradiated with 0–20 Gy of x-ray, replated, and cultured for 24 h before fixation in 70% ethanol. Cell were then stained with anti-BrdU Ab (BD Pharmingen) as described previously (19).

Purification and in vitro activation of B cells

Enriched populations of splenic B cells were obtained using magnetic anti-CD19 beads according to the manufacturer’s instructions (BD Pharmingen). Cells (2 × 10⁵) were plated in 24-well flat-bottom plates in 1.0 ml of complete medium consisting of RPMI 1640 (Invitrogen Life Technologies) supplemented with 10% FCS, 1% sodium pyruvate, 1% nonessential amino acids, 0.5% l-glutamine, 5 × 10⁻⁵ M 2-ME, 100 U/ml penicillin, and 100 μg/ml streptomycin, and incubated at 37°C in 5% CO₂. To induce specific Ig class switching, B cells were stimulated as follows: for IgG1, IgG2a, and IgM, the stimulus was 2.15 μg/ml mouse membrane CD40L (PeproTech) plus 1,000 U/ml recombinant mouse IL-4 (PeproTech); for IgG2b and IgG3, the stimulus was 15 μg/ml LPS (Sigma-Aldrich) as described previously (18). Culture supernatants were harvested on day 6 for quantitation of Ig classes by ELISA as described above.

Digestion-circularization-PCR (DC-PCR)

Genomic DNA (1 μg) isolated from cultured B cells was digested by EcoRI (Nippon Gene) overnight, and 100 ng was self-ligated overnight using T4 DNA ligase (TAKARA) in a 50 μl volume as described previously (18). Ligated DNA was subjected to two rounds of PCR using nested primer pairs for DC-PCR of Sμ-Sγ1, Sμ-Sε, and a nicotinic acetylcholine receptor (nAChR) as a control. Primer sequences are as follows: Sμ-Sγ1 1st forward, 5′-gagcagctaccaaggatcaggga-3′ and Sμ-Sγ1 1st reverse, 5′-cttcacgccactgactgactgag-3′; Sμ-Sε1 1st forward, 5′-cgggtattggaaaataattgaatg-3′ and Sμ-Sε1 1st reverse, 5′-gcagagcatcctcacataca-3′; nAChR 1st forward, 5′-gcaaacagggctggatgaggctg-3′ and nAChR 1st reverse, 5′-gtcccatacttagaaccccagcg-3′; Sμ-Sγ1 2nd forward, 5′-ggagaccaataatcagagggaag-3′ and Sμ-Sγ1 2nd reverse, 5′-gagagcagggtctcctgggtagg-3′; Sμ-Sε1 2nd forward, 5′-gtccttcaatttcttacataacc-3′ and Sμ-Sε1 2nd reverse, 5′-atgcaggatagaccccagac-3′; nAChR 2nd forward, 5′-ggactgctgtgggtttcacccag-3′ and nAChR 2nd reverse, 5′-gccttgct tgcttaagaccctgg-3′.

Irradiation experiments

Eight-week-old mice were irradiated in an x-ray irradiator (Hitachi; MBR-1520R-3) at a rate of 0.5 Gy/min in a single dose of 8 Gy, and survival was determined.

Analysis of intracellular ROS

Cells were loaded with 5 μM 2′-7′-dichlorofluorescene diacetate (DCF-DA) (Sigma-Aldrich) and incubated on a shaker at 37°C for 30 min (14). The peak excitation wavelength for oxidized DCF was 488 nm and emission was 525 nm.

PCR analysis

For RT-PCR, analysis of germline and productive IgG1 and IgE transcripts in B cells were activated in vitro as described previously (18). The cDNA from total RNA was used in PCR. Primer sequences were as follows: germline Igγ1 forward, 5′-tcgagaagcctgaggaatgt-3′ and germline Igγ1 reverse, 5′-atagacagatgggggtgtcg-3′; germline Igε forward, 5′-ctggccagccactcacttat-3′ and germline Igε reverse, 5′-cagtgcctttacagggcttc-3′; productive Igγ1 forward, 5′-tcgagaagcctgaggaatgt-3′ and productive Igγ1 reverse, 5′-atagacagatgggggtgtcg-3′; productive Igε forward, 5′-ttggactactggggtcaagg-3′ and productive Igε reverse, 5′-cagtgcctttacagggcttc-3′. For intra- and interchromosomal rearrangements, genomic DNA (100 ng) derived from (wild-type) WT and Atm^−/− thymocytes was used for PCR amplification. The PCR conditions were essentially as described in Lista et al. (20). Primer sequences are were follows: TCRGV3S1a, 5′-accatacactggtaccggca-3′; TCRGJ1/2a, 5′-tcatcactggaataaagcag-3′; TCRBV5S1b, 5′-ttctcagtccaacagtttgat-3′; TCRGV3S1b, 5′-acccctacccatattttcttag-3′; TCRBJ2a, 5′-tctactccaaactactccag-3′; TCRBV5S1a, 5′-tggtatcaacagactcagggg-3′; TCRBJ2b, 5′-cctccaagcgaggagatgtgaa-3′; TCRGJ1/2b, 5′-ggtacttaccggagggaatt-3′.

Switch junction analysis

Sμ-S1 junctions were amplified from genomic DNA as described previously (18).

Karyotyping analysis

MEFs at passage 6 were arrested at mitosis by treatment with 0.002 μg/ml colcemid for 2 h. Metaphase spreads were prepared by standard procedures and stained with 4′,6′-diamidino-2-phenylindole, Hoechst 33258, and quinacrine mustard for identification of individual chromosomes. Over 100 cells were subjected to analysis.

Statistical analyses

The p values for most experiments were calculated using the unpaired Student’s t test. Statistical differences in Kaplan-Meier survival curves were determined using the log-rank rest (p < 0.005).

NAC inhibition of premature senescence in Atm^−/− MEFs

To evaluate base levels of ROS in cells of Atm^−/− mice, we measured H₂O₂ concentrations in MEFs derived from WT and Atm^−/− mice. MEFs from untreated Atm^−/− mice showed a much higher base level of H₂O₂ than did WT MEFs (Fig. 1⇓A). NAC treatment did not affect WT cells but did dramatically reduce ROS levels in Atm^−/− MEFs. We then evaluated the effects of NAC on MEF proliferation. Although MEFs derived from WT E14.5 embryos proliferate rapidly at early passages, the growth rate slows over 10 passages such that cells eventually stop dividing. Untreated Atm^−/− MEFs showed no defects in proliferation at the early stages (passage 2), grew at a reduced rate at later passages (passage 6), and underwent complete growth arrest by passage 10 (Fig. 1⇓B). However, when proliferating Atm^−/− MEFs were treated with NAC, they maintained WT rates of proliferation at passages 2 and 6 and showed a significant proliferation even at passage 10. Abnormalities in cell morphology were also mitigated by NAC. By passage 10, untreated WT MEFs exhibit elongated, spindle-shaped, fibroblastic morphology, whereas untreated Atm^−/− MEFs show a refractile cytoplasm with flat, enlarged morphology. Unlike WT MEFs, untreated Atm^−/− MEFs also stained positively for a senescence-associated β-galactosidase (SA-β-gal) (Fig. 1⇓C). NAC treatment prevented the appearance of these senescence-associated phenotypes in late passage Atm^−/− MEFs. Karyotypic analysis of MEFs at passage 6 revealed that the frequency of chromosomal aberrations was higher in untreated Atm^−/− MEFs than in WT MEFs, as reported previously (21). Importantly, NAC treatment prevented chromosomal abnormalities in Atm^−/− MEFs (Fig. 1⇓D). However, NAC treatment could not rescue the impaired G₁/S checkpoint seen in Atm^−/− MEFs (Fig. 1⇓E). These results indicate that chromosomal instability seen in Atm^−/− MEFs, but not the defective G₁/S checkpoint, is caused by ROS elevation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Prevention of premature senescence in Atm^−/− MEFs by NAC. A, Reduction in H₂O₂ levels. Cultured WT and Atm^−/− MEFs were analyzed for levels of intracellular H₂O₂ as determined by DCF-DA. H₂O₂ concentrations in Atm^−/− MEFs were restored to WT levels by treatment with 100 μM NAC. Results are expressed as the mean ± SD of the ratio of DCF fluorescence of MEFs to those in nontreated WT MEFs (n = 3). B, Prevention of growth arrest. WT and Atm^−/− MEFs were maintained in replicate cultures with or without NAC, and cell numbers were determined using a Coulter counting on the indicated days. Proliferation curves are shown at passages 2, 6, and 10. Results are the mean cell number ± SD of five plates for each time point. C, Decrease in SA-β-gal-positive cells. Cultures of WT and Atm^−/− MEFs, with or without NAC, were examined for the presence of SA-β-gal-positive cells (indicating premature senescence). Left panel, SA-β-gal staining; right panel, quantitation of data. Results are the mean number ± SD of SA-β-gal-positive cells per high-power field (HPF) (n = 3). D, Reduced frequency of chromosomal instability. WT and Atm^−/− MEFs at passage 6 were treated with or without NAC, and metaphase spreads were assessed for the frequency of chromosome aberrations. Data shown are the percentage of cells in a culture showing structural abnormality and are representative of three independent experiments. E, Lack of rescue of the G₁/S checkpoint in Atm^−/− MEFs by NAC. Passage 2 WT and Atm^−/− MEFs, with or without NAC, were synchronized in G₀ by serum starvation, exposed to IR, and incubated with BrdU. Entry into S phase was evaluated by flow cytometry using anti-BrdU Ab. Left panel, mean percentage ± SD of BrdU-positive cells (expressed as the percentage of control nonirradiated cells) 24 h after the indicated dose of IR (n = 3). Right panel, Mean percentage ± SD of BrdU-positive cells at indicated time points after exposure to 20 Gy IR (n = 3).

Inhibition of acute sensitivity of Atm^−/− mice to irradiation by NAC treatment

Next, we investigated roles of ROS regulation on phenotypes of Atm^−/− mice in vivo. To do so, NAC was administered in the drinking water of Atm^−/− mice. Untreated Atm^−/− mice weighed significantly less than WT mice, as reported previously (6, 7, 8). NAC treatment did not affect somatic growth of Atm^−/− mice (data not shown). It was reported previously that Atm^−/− mice showed hypersensitivity to total body irradiation (6), indicating that Atm loss leads to an abnormal response to DNA damage. When adult mice were subjected to IR treatment at the level of 8 Gy, all Atm^−/− mice died within a week, whereas WT mice died slowly, as previously reported (6) (Fig. 2⇓A). However, when mice were treated with NAC in vivo, the early death of Atm^−/− mice by IR was significantly inhibited. The death of Atm^−/− mice by irradiation resulted from acute radiation toxicity to the gastrointestinal tract. Epithelial crypt degeneration observed in the small intestine of these mice was inhibited by NAC treatment (Fig. 2⇓, B–D). These data demonstrate that abnormal in vivo responses to IR seen in Atm^−/− mice can be rescued by ROS suppression.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Inhibition by NAC treatment of acute sensitivity of Atm^−/− mice to IR. A, Survival curves of WT and Atm^−/− mice with or without in vivo treatment of NAC after 8 Gy of x-ray irradiation (n = 6 each group). The difference between Atm^−/− mice plus and minus NAC is statistically significant (Wilcoxon two-sample test assuming nonparametric data and a two-tail p value, p < 0.05). B and C, H&E-stained sections of the small intestine taken at day 4 after 8 Gy irradiation are shown. WT (B), Atm^−/− (C), Atm^−/− with NAC (D).

Prevention of lymphomagenesis in Atm^−/− mice by NAC

Schubert et al. (15) and Reliene et al. (22) reported that treatment of Atm^−/− mice with the antioxidants extend their lifespan by delaying the onset of thymic lymphoma. To confirm the role of elevated ROS in Atm-related lymphomagenesis, we administered NAC in drinking water from birth to WT and Atm^−/− mice. We confirmed that NAC treatment reduced ROS levels in lymphocytes in vivo (Fig. 3⇓). As expected, untreated Atm^−/− mice began dying of lymphoma at 10 wk of age, and only 30% survived until 24 wk (Fig. 4⇓). Strikingly, NAC treatment significantly prevented lymphomagenesis in Atm^−/− mice.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Reduced ROS in vivo in Atm^−/− mice with NAC treatment. Restored elevation of ROS in Atm^−/− thymocytes (A) and B cells (B) by NAC treatment. H₂O₂ levels were determined by DCF-DA staining in freshly isolated thymocytes or CD19⁺ cells (B cells) from spleen of WT and Atm^−/− mice, with or without NAC treatment. Results are expressed as number of fold increase in intracellular ROS compared with WT cells without NAC treatment (n = 3). ∗, p < 0.01.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Prevention of tumorigenesis in Atm^−/− mice by NAC. Kaplan-Meier survival curves are shown for WT and Atm^−/− mice (n = 16 each group) administered 1 mg/ml NAC in their drinking water.

Restoration of impaired CSR in Atm^−/− B cells by NAC

These data prompted us to investigate effects of NAC on the development of pretumor lymphocytes in Atm^−/− mice. It has recently been reported that Atm is required for efficient recombination between class switch regions of Ig genes, a process requiring regulated DSB repair (18, 23). We therefore analyzed the effect of ROS reduction on defective CSR seen in Atm^−/− mice. CSR is required for production of IgG but not IgM. Accordingly, and consistent with previous reports (18, 23), levels of serum IgG1, IgG2a, and IgG3 in untreated Atm^−/− mice were markedly lower than those in WT littermates (Fig. 5⇓A). However, in Atm^−/− mice, NAC treatment in vivo dramatically restored serum IgG concentrations to WT levels. To investigate the intrinsic ability of Atm^−/− B cells to undergo CSR in vitro, we examined the effect of NAC on in vitro secretion of Igs by splenic B cells cultured with various polyclonal B cell activators. Again, the IgM response of Atm^−/− B cells did not differ significantly from that of WT B cells, with or without NAC treatment (Fig. 5⇓B). However, secretion of various IgG classes by Atm^−/− B cells was impaired, and this defect was rescued by NAC treatment. In addition, NAC treatment restored surface IgG1 expression to Atm^−/− B cells (Fig. 5⇓C). There were no differences in in vitro clonal expansion of WT and Atm^−/− B cells, with or without NAC (data not shown). These data demonstrate that ROS elevation in Atm^−/− B cells results in an intrinsic defect in CSR in B cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Restoration of impaired CSR in Atm^−/− B cells by NAC. A, Rescue of specific Ig classes in serum. Serum samples from WT and Atm^−/− mice, treated with or without NAC, were assayed for levels of specific Ig classes by ELISA. Data shown are the mean Atm^−/− titer ± SD calculated as a percentage of the WT titer (n = 3). □, Serum from untreated Atm^−/− mice; ▪, serum from NAC-treated Atm^−/− mice. B, Rescue of specific Ig classes in B cell culture supernatants. CD19⁺ B cells were isolated from spleens of WT and Atm^−/− mice treated with or without NAC. B cells were stimulated in vitro as described in Materials and Methods, and levels of specific Ig classes in culture supernatants were analyzed by ELISA. Data shown are the mean Atm^−/− titer ± SD calculated as a percentage of the WT titer (n = 3). □, Atm^−/− without NAC; ▪, Atm^−/− with NAC. C, Rescue of surface IgG1 expression on Atm^−/− B cells. Cultured B cells in B were harvested after 4 days of in vitro stimulation with CD40L and IL-4, and surface IgG1 was determined by flow cytometry. Numbers shown are the percentage of B cells that had undergone CSR expressed as a proportion of total B220⁺ B cells. D, Rescue of germline and productive Ig transcripts. Gene expression profiles of WT and Atm^−/− B cells stimulated in vitro (as in B) were determined by RT-PCR analysis of total RNA. Germline and productive IgG1 (γ1) and IgE (ε) transcripts were measured on poststimulation days 3 and 6, respectively. Data shown are one trial representative of three independent experiments. D.D.W., Double-distilled water, G3PDH (PCR control). E, Rescue of CSR. Genomic DNA was isolated from WT and Atm^−/− B cells activated in vitro for 6 days and analyzed for CSR from IgM to IgG1, and from IgM to IgE, by DC-PCR as described in Materials and Methods. PCR was performed using primers specific for the indicated recombined switch regions. nAChR, DC-PCR control. Data shown are one trial representative of three independent experiments.

We next examined expression of germline (Gγ1, Gε) and productive (Pγ1, Pε) transcripts of Ig genes using RT-PCR. Germline and productive Ig transcripts appear before and after CSR, respectively. As expected, induction of Gγ1 and Gε transcripts peaked on day 3 in both WT and Atm^−/− B cell cultures, with or without NAC (Fig. 5⇑D). However, expression of Pγ1 and Pε transcripts in Atm^−/− B cells was markedly reduced compared with WT cells in the absence of NAC. Upon NAC treatment, expression of productive transcripts in Atm^−/− B cells was restored. The PCR results were confirmed by a semiquantitative method (data not shown). To determine directly whether NAC treatment improves class switching by acting at the level of DNA recombination, we performed a DC-PCR assay to measure CSR directly in B cells. In B cells with normal CSR, the 5′ switch region DNA sequence for IgM (Sμ) is contiguous with the 3′ switch region for IgG1 (Sγ1) or the switch region for IgE (Sε) at the recombined IgH locus. To detect these fragments when Sμ and Sγ1, or Sμ and Sε, were present on the same DNA fragment, we amplified the switch regions using DC-PCR. This analysis revealed that Atm^−/− B cells treated with NAC generated WT levels of both Sμ-Sγ1 and Sμ-Sε fragments, whereas CSR in untreated Atm^−/− B cells was significantly lower than that seen in WT cells (Fig. 5⇑E). The PCR results were confirmed by a semiquantitative method (data not shown). These data demonstrate that regulation of ROS by Atm is required for efficient DNA recombination.

Prevention of defective T cell development and aberrant TCR rearrangement in Atm^−/− mice by NAC

We next determined the effects of NAC on T cell development in Atm^−/− mice. Thymic cellularity in untreated Atm^−/− mice (n = 5) was 41 ± 15% of that seen in WT animals. However, NAC-treated Atm^−/− mice retained 83 ± 21% of WT thymic cellularity, suggesting a partial rescue. Flow cytometric analysis revealed a significant reduction in the percentage of CD4⁺CD8⁻ T cells in the Atm^−/− thymus (Table I⇓ and Fig. 6⇓A), indicating that Atm is essential for the differentiation of CD4⁺CD8⁻ T cells from CD4⁺CD8⁺ thymocytes. With NAC treatment, the percentage of CD4⁺CD8⁻ T cells in the Atm^−/− thymus was comparable to that seen in thymi of untreated or NAC-treated WT mice. Although the proportion of Thy-1⁺ cells was relatively normal in the Atm^−/− thymus, the loss of Atm resulted in a striking reduction in cells expressing either TCRαβ or CD3, an outcome reversed by NAC treatment (Table I⇓ and Fig. 6⇓, B and C). NAC treatment also normalized T cell populations in the Atm^−/− spleen (Table I⇓ and Fig. 6⇓D). Thus, NAC treatment restored T cell development in Atm^−/− mice. We then investigated chromosomal translocation seen in T cells from Atm^−/− mice: an interchromosomal rearrangement between the TCRβ and TCRγ genes that predicts the onset of global chromosomal translocations in mouse thymocytes (20). TCRβ-TCRγ exchange was observed in thymocytes of Atm^−/− mice (Fig. 7⇓). However, only a faint band of interchromosomal rearrangement was observed in thymocytes from NAC-treated Atm^−/− mice, indicating that ROS elevation induces aberrant V(D)J recombination. These data demonstrate that Atm-mediated ROS regulation is essential for proper DNA recombination, preventing immunodeficiency and tumorigenesis.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Restoration by NAC of T cell development in Atm^−/− mice. Thymocytes or splenic T cells were isolated from WT and Atm^−/− mice treated with or without NAC in vivo. Flow cytometry was performed to assess expression of CD4/CD8 (A) and Thy1/TCR (B) by thymocytes, of CD3 by thymocytes (C), and of CD4/8 by splenocytes (D). Numbers in quadrants are the percentage of total cells expressing the indicated markers. In each case, one trial representative of 10 independent experiments is shown.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Prevention by NAC of aberrant TCR rearrangement in Atm^−/− mice. Intra- and interchromosomal rearrangements were monitored by PCR in precancerous thymocytes from WT and Atm^−/− mice treated with or without NAC in vivo. A, A schematic diagram of intra- and interchromosomal rearrangements in the TCRβ and TCRγ loci is shown in the upper panel. The TCRγV3S1TCRβJ2 rearrangement represents formation of a t (6 ,13 ) chromosomal translocation. B, Shown is representative genotypic PCR analysis of rearrangements shown in A in untreated and NAC-treated WT and Atm^−/− cells. Atm WT and mutant alleles are also shown.